Hybrid detection method for spectrum sensing in cognitive radio

Since the spectrum is the scarce resource, spectrum sensing is an important part of Green Engineering. The cognitive radio has been proposed as the future technology to meet the ever increasing demand of the radio spectrum by allocating the spectrum dynamically to allow unlicensed access on non-interfering basis. In cognitive radio, spectrum sensing is the fundamental problem. Energy detection method is the simplest method but suffers from noise uncertainty problem. Covariance based detection exploits space-time signal correlation, but its performance is very sensitive to the signal correlation. Hybrid detection method exploits the advantages of these two methods by utilizing energy detection in low correlation and covariance based detection in high correlation. We analyze the performance of hybrid detection method exploiting TV White Space under Rayleigh and Rician fading channel for different Signal-to-Noise Ratio values by setting probabilities of false alarm and measuring probability of detection. Simulation shows that the hybrid detection method outperformed energy detection and covariance based detection method and is more insensitive to the type of input data. The hybrid detection method shows the better performance compared to energy detection and covariance method under time-varying fading conditions.